Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Encoder improvements #1254

Closed
dennisbader opened this issue Oct 3, 2022 · 1 comment
Closed

Encoder improvements #1254

dennisbader opened this issue Oct 3, 2022 · 1 comment
Assignees
@dennisbader
Labels
improvement New feature or improvement

Comments

@dennisbader
Copy link
Collaborator

Improve encoders:

  • in addition to input_chunk_length and output_chunk_length, add a max_past_lag and max_future_lag parameter to encoder creation. With this, we could move the logic that is now implemented in the forecasting models to the Encoder side (for example RegressionModel._model_encoder_settings() and RegressionModel._get_encoders_n() ). This would make it easier to use the encoders as standalone covariate generators
  • add add_encoders to all models (LFMs still don’t support it)
  • rename future_covariate and others covariate params to future_covariates for consistency (breaking for users that use encoders as standalone but I think it makes sense to rename them)
  • move encoder module from darts.utils.data into darts.dataprocessing.encoders ?
@dennisbader dennisbader added the improvement New feature or improvement label Oct 3, 2022
@dennisbader dennisbader self-assigned this Oct 3, 2022
@eliane-maalouf eliane-maalouf mentioned this issue Oct 11, 2022
Open
@dennisbader
Copy link
Collaborator Author

Closing as this was added in #1338

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@dennisbader dennisbader

Labels
improvement New feature or improvement
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@dennisbader